Epistemologia da virtude - vol. I: Crença apta e conhecimento reflexivo

by Ernest Sosa, Luiz Paulo Rouanet (Translator)

Epistemologia da virtude apresenta nova abordagem para alguns dos mais antigos e interessantes problemas da filosofia, os do conhecimento e do ceticismo. Ernest Sosa argumenta em favor de dois níveis de conhecimento, o animal e o reflexivo, cada um visto como uma capacidade humana distinta. Ao adotar um tipo de epistemologia da virtude compatível com a tradição encontrada em Aristóteles, Tomás de Aquino, Reid e, especialmente, Descartes, ele apresenta uma explicação do conhecimento que pode ser utilizada para iluminar certas variedades de ceticismo, sobre a natureza e o estatuto das intuições e a normatividade epistêmica.

Genres: Philosophy

144 pages, Paperback

First published June 28, 2007

Reviews

[Petra · Rating 4 out of 5]

Four stars.

A very interesting introduction to Virtue Epistemology, which, despite my predilection for virtue-based approaches in general, I had somehow never touched before now. Thankfully, on top of giving me a rather good primer, Sosa has given me an extensive bibliography to devour.

The main negative point I have for Sosa's lecture series here is that he feels a little unfocused. Sometimes this happens directly in terms of tangents, but more often it's the frameworks and issues he chooses to frame, which feel slapdash and all over the place. Though, this could also just be that I'm well out of touch on modern discussions (my experience, to the extent it can be called this, typically ends around 1950).

So what is Virtue Epistemology?

The core idea is very very simple: Knowledge is True Belief that is Apt to the knower. That is to say, knowledge is an intellectual achievement: true belief that is achieved through the competence of the knower, in whatever relevant area(s).

This simple little idea has many virtues:

First and most simply, it's generous without being infantilising. This is not the most important reason by far, but I find it pleasant when what's good is good and is recognised as such. When people are noted and valued for succeeding, even in a pedestrian way. This seems to me to be a broadly healthy outlook.

Second, it does surprisingly well against the competition, handily cutting through the usual opponents: skepticism and Gettier problems. But first, some terms:

Safety and Security:

Safety is when a belief (or performance generally) could not easily have been false or a failure. It is strong, reliable if you like (though reliable is also a technical term in the literature). Security on the other hand is when were something false or wrong, it would be known to be.

Aptness:

A performance is apt when it is successful, and this success is due to a relevant competence on the part of the performer, utilised in the appropriate/normal circumstance. I could pick up a bow and arrow and by chance hit a target due to freak winds tomorrow, doesn't mean I would be a good archer. The shot would be good, but not apt: it does not really belong to me. A good archer makes apt shots: shots that are successful due to their repeatable internal competence in archery.

Note that the good archer does not need to be perfect, nor is archery free from interference from outside mechanisms: a good archer can make a shot that would be great, but the same freak wind that carried my shot to the target may blow theirs off course. There is a whole Sorites Paradox about how much external influence can be present before a performance stops being apt, but that's neither here nor there.

Note that the aptness of a performance is orthogonal to both safety and security:

Firstly, the appropriate or normal conditions which would be necessary for aptness could fail to obtain: the good archer's gust of wind example above for instance. Thus aptness itself is not safe.

Secondly, re: security, a performer could fail to detect that the appropriate conditions do not apply.

Thirdly, the good archer can just miss the shot sometimes. "Could not easily be false" =/= "could not possibly be false." In such a case a performance may be safe yet unsuccessful and therefore not apt.

Animal versus Reflective Knowledge:

More or less does what it says on the tin: animal knowledge is apt belief (successful because competent), reflective knowledge is apt belief aptly considered. In other words, reflective knowledge is successful introspective reflection upon the deliverances of our senses or our intellectual intuitions. A correct understanding of its standing both within us and its relation (true, false, more complicated...) to the external world.

Epistemic Normativity:

We are beings who judge, who make decisions and evaluate. We, a homo aestimans, do this with absolutely everything, in every domain of human activity and thought. In epistemology, science and arguably philosophy in general, the fundamental value is Truth. And Sosa is right that this is an evaluative process: if we didn't think Truth was worth searching for, we wouldn't do any of these things, we wouldn't even have language.

Given truth as the fundamental value and telos in this domain, everything else is measured by this yardstick. Just as a tennis racket is good or bad only relative to its ability to perform its role (in fact the racket is entirely pointless outside of this field), so too is a certain thought form, evaluative practice, scientific methodology good or bad in light of its (in)ability to deliver truth.

As with many things in virtue-based approaches, this is simultaneously blindingly obvious once stated, and a new and useful framing to a discipline recovering from terminal Reddit-Brain. It is another tool in leading epistemology away from the arid badlands of positivism and its ilk and towards a regrounding of the human--the good and the bad--in the Wissenschaftslehre.

Everything that we do and are is a process and a practice, and Virtue Epistemology not only reflects this fact in its specific domain, it also helps ground philosophy and science in the domain of human social practice. It can help us wrap the Human Project up in an iridescent, teleological cloak with is both invigorating and makes for good narrative.

Sosa adds to this that apt belief should be considered the fundament rather than truth in and of itself, mostly due to Gettier problems both modern and ancient: if one hits upon the truth by chance either ex nihilo or because of a deformed procedure, one is more lucky than capable. I don't know if this is really the distinction that Sosa says it is however; all epistemic aptness is is the possession of success (knowledge) through relevant skill, but I fail to see how "good because skilled at getting truth" does not rely on a prior "truth=good." I admit this is a pedantic point, but to me, rather than the fundament, epistemic aptness is the first emanation from the fundament for the purpose of us as finite beings.

Sosa's Argument:

With that out of the way, Sosa takes Virtue Epistemology through the usual shakedown cruise of fighting skepticism, showing that skeptics have flawed concepts of knowledge, based on criteria of security rather than safety, then explains intuition through a Virtue-based framework. A lot of this section feels unnecessarily introductory, but he gets the job done.

Virtue-based approaches have the benefit of tackling typical issues in epistemology head-on: fallacious reasoning for instance is "a performance error chargeable against the subject." Everything is capabilities. By getting into the muck from the get-go, we can naturalise our fallible intuitions and attractions without making weird reifications:

By analogy to the seemings delivered by our visual system, the intuitions immediately delivered by our rational competences are preponderantly true, even if occasionally false. This is why those rational mechanisms are intellectual competences, because they systematically lead us aright. All seemings delivered by such competences are thereby epistemically justified.

Animal Knowledge and Gettier Cases:

So the level 1 enemy of skepticism has been vanquished, intuition has been situated, and we're left with a safety criterion over a security criterion. Except even this is too strong, Sosa argues. As he puts it,

The fact that the thermostat fails when disconnected is not relevant to whether it is a good thermostat. Nor is it relevant to its performing well that it might very easily have been disconnected...Suppose that it might very easily have been disabled [by glue], or out of position for proper operation; nevertheless, so long as it was actually in working order, and normally connected, the device's performance is apt even if unsafe, and creditable to it as its doing.

Which, I mean, fair enough. This does a good job of cutting up skeptical arguments and revealing them as stupid. Rephrased, it sounds like something an annoying teenager would come up with: "you say that you know that the sky is blue, but you wouldn't know anything if I shot you in the head!" Like yeah, congratulations on your powers of deduction Sherlock. My existence is finite and things could have been different.

Sosa also gets into Gettier cases to further his argument. The case he refers back to the most is the Kaleidoscope Perceiver, which is really about establishing the status of animal knowledge for sensory perception.

The Kaleidoscope Perceiver:

You're in a dark room, and there is a window on the wall into an exhibit. This window leads into a closed box of its own, its walls are black, and there is a single object in the box: a ball. The ball looks red to you, but the light source is out of view. You conclude that the ball is red, but is it really? After all, it could be a red light source and a white ball.

Of course, for this to be a Gettier case, the ball actually is red, the problem of course being whether and to what extent you actually know that the ball is red, and what this implies for perception more broadly. Sosa goes on to argue that the Kaleidoscope Perceiver (KP) has animal knowledge but not reflective knowledge. But does this hold up?

First of all, given the ridiculous setup required to make this viable, normal conditions obviously do not hold. In any normal situation, an ably-sighted person would correctly attribute the colour. This seems to undermine Sosa's claim: given that aptness simply doesn't apply in this situation, how can one have apt belief? It seems a category error.

The second point is that there is a very valid question over what the KP actually assumes in this situation. There is a big difference between "this ball appears red to me" (undeniably true), and "this ball appears red and is therefore red in itself and therefore painted red or made of a red substance." This is the method by which Sosa wriggles out of the conclusion of the previous paragraph: the KP has animal knowledge that the ball appears red, but lacks reflective knowledge of the actual nature of the ball. If the KP catches that hijinks are afoot, then the only answer they can give to the actual colour of the ball is "I don't know," and if they don't catch it, then they're making a false inference. Either way, they lack reflective knowledge.

As much as this intuitively feels like a cop-out, it has a certain strength. After all, the KP is not the same as someone hallucinating the redness of a ball, there is some truth here, they aren't making up the redness of whole cloth.

This is the problem with the thrust of the skeptic: by being so all-or-nothing, he reduces a complex situation like the KP, and even our regular common-sense knowledge to that of the hallucinator, when that simply isn't what's happening. The KP knows something about the ball, they just don't know the why. In common, successful perceptual knowledge we know both our direct perception and the why and therefore have both animal and reflective knowledge.

Likewise, usually perceiving a red-looking object means that the surface is in itself red, and the fact that this weird situation has been engineered does not undermine the broad reliability of the inference in general use.

Conclusion:

Just like G.E. Moore, or Roy Bhaskar, or even Diogenes, what Sosa is doing is bringing the cumbersome but awe-inspiring artillery of common sense to bear on the increasingly cloistered and self-referential scribblings of the competition.

I like it, it's good. It works.

[Pavlogreen]

(Leído en la versión española de Juan J. Colomina Almiñana, Una epistemología de virtudes).

Otra obra más de Sosa. A medida que he ido leyendo textos de este autor me he ido dando cuenta de su loable capacidad para repetirse y decir prácticamente lo mismo, con matices, pequeñas adiciones y ejemplos nuevos, en todos sus libros. Este, quizá el más famoso de los suyos, me demuestra esto con creces. Otro de los detalles de los que me dado cuenta cuanto más leía a Sosa es que su idea de la epistemología resulta estrecha y se deja sin abarcar partes importantes de una teoría del conocimiento que se precie, como una teoría de la verdad o una caracterización sólida del impacto social de conocimiento, entre otras. Hay que reconocer que los acotados terrenos a los que Sosa consagra su trabajo son muy brillantes y su enfoque de virtudes se ha vuelto casi imprescindible para hilvanar una epistemología en este siglo. Recojo su guante en este aspecto, pero me permito resaltar lo insuficiente que resulta su filosofía, no por un afán destructivo, sino más bien con la esperanza de enriquecer sus excelentes cimientos, que sin embargo no dejan de ser solo eso, cimientos.

[Edmundo · Rating 5 out of 5]

Wonderful book!